Lake eutrophication management modeling using dynamic programming

Lake eutrophication problems have received considerable attention in Taiwan, especially because they relate to the quality of drinking water. In this study, steady-state river water quality and lake eutrophication models are solved using dynamic programming algorithms to find the nutrient removal rates for eutrophication control during dry season. The kinetic cycle of chlorophyll-a, phosphorus and nitrogen for a complete-mixed lake is considered in the optimization framework. The Newton-iterative technique is adopted to solve the nonlinear equations for the steady-state lake eutrophication model. The optimization framework is applied to Cheng-Ching Lake in southern Taiwan. Several nutrient loading scenarios for eutrophication control are studied. Optimization results for nutrient removal rates and corresponding wastewater treatment capacities of each reach of the Kao-Ping River define the least cost approach to lake eutrophication control. A natural purification method, structural free water surface wetland, is also suggested to save more investment and improve river water quality at the same time.     

Factors affecting the concentrations of PM10 in central Taiwan

In this study, the synoptic weather types that have high probability and low probability of producing PM10 episode are referred to as HPE and LPE, respectively. Multiple linear regressions analysis showed that NO2 was the most important contributor (35.61%) to the concentrations of PM10 for HPE weather. For LPE weather, the season factor had the greatest contribution (48.11%) to the concentrations of PM10. Using the correlation coefficients between the concentrations of PM10 and SO2 or NO2 on HPE and LPE to calculate the increase of PM10 from LPE to HPE, we found that the increase of PM10 owing to the increase of SO2 and NO2 from LPE to HPE was 12.93microg/m3 which was about 51% of the total amount of PM10 increased from LPE to HPE. Results of factor analysis showed that the first component could be attributed to the result of local pollution especially for the weather patterns of types P3 and P6, while the secondary component for the weather patterns of types P1 and P4 can be attributed to the long-range transport of SO2 pollutants from China.     

Valuing the benefits of the urban forest: a spatial hedonic approach

This paper measures the benefits of the urban forest by examining its effect on housing prices. A Geographic Information System is used to develop a measure of the urban forest, the Normalised Difference Vegetation Index, from satellite imagery and to construct other variables from a variety of sources. Spatial hedonic housing price models for the Indianapolis/Marion County area are estimated. The models indicate that greener vegetation around a property has a positive, significant effect on housing price, holding everything else constant. This effect is dominated by measures at the neighborhood level. These findings indicate that property owners value the urban forest, at least in part, by the premium they pay to live in neighborhoods with greener, denser vegetation. These findings also indicate that public action to maintain and enhance the urban forest may be warranted. Planners and urban foresters can use these findings to inform public and policy debates over urban forestry programs and proposals.     

Forecasting relative impacts of land use on anadromous fish habitat to guide conservation planning.

Land use change can adversely affect water quality and freshwater ecosystems, yet our ability to predict how systems will respond to different land uses, particularly rural-residential development, is limited by data availability and our understanding of biophysical thresholds. In this study, we use spatially explicit parcel-level data to examine the influence of land use (including urban, rural-residential, and vineyard) on salmon spawning substrate quality in tributaries of the Russian River in California. We develop a land use change model to forecast the probability of losses in high-quality spawning habitat and recommend priority areas for incentive-based land conservation efforts. Ordinal logistic regression results indicate that all three land use types were negatively associated with spawning substrate quality, with urban development having the largest marginal impact. For two reasons, however, forecasted rural-residential and vineyard development have much larger influences on decreasing spawning substrate quality relative to urban development. First, the land use change model estimates 10 times greater land use conversion to both rural-residential and vineyard compared to urban. Second, forecasted urban development is concentrated in the most developed watersheds, which already have poor spawning substrate quality, such that the marginal response to future urban development is less significant. To meet the goals of protecting salmonid spawning habitat and optimizing investments in salmon recovery, we suggest investing in watersheds where future rural-residential development and vineyards threaten high-quality fish habitat, rather than the most developed watersheds, where land values are higher.